Professional AI Agents & Skills for Claude Code | Covering Development, DevOps, Cloud, and Beyond
Important
π This project will be part of a bigger ecosystem! STAY TUNED
π₯ Your contributions will be recognized and featured with your name and GitHub profile.
π Join our community of developers making AI coding better for everyone!
FastAgent is a curated collection of professional plugins for Claude Code that extend your AI coding assistant with specialized agents, skills, and commands. Our approach is simple: pick only what you need, avoid token bloat, and get better results.
- π― Focused & Efficient: Each plugin is laser-focused on specific technologies or roles
- π° Token-Conscious: Only load what you need - save tokens, save costs
- π§ Modular Design: Mix and match plugins based on your workflow
- π¦ Easy Installation: One command to add capabilities
- π Production-Ready: Battle-tested patterns and best practices
We organize plugins into three tiers to make it easy to choose what you need:
Broad-spectrum plugins that benefit most developers. These cover fundamental workflows like code review, debugging, and refactoring.
Who should use: Everyone
Role-based plugins combining multiple technologies and workflows for specific job functions.
Who should use: Developers in specialized roles
Technology-specific plugins for deep expertise in particular tools and frameworks. Choose the ones matching your stack.
Who should use: Developers working with specific technologies
Make sure you have Claude Code installed.
# In Claude Code, run:
/plugin marketplace add armanzeroeight/agents-skills# Open the plugin browser
/plugin
# Or install directly
/plugin install <plugin-name>@fastagent-marketplaceOnce installed, agents and skills activate automatically based on your work context. Commands are available via slash commands.
| Plugin | Description | Agents | Skills | Commands |
|---|---|---|---|---|
| testing-toolkit | Comprehensive testing toolkit for test strategy, coverage analysis, and test data generation | Test Strategist | Test Coverage Analyzer, Test Data Generator | /run-tests, /coverage-report |
| git-workflow-toolkit | Git workflow and branching strategy toolkit with commit message generation and branch management | Git Expert | Commit Message Generator, Branch Strategy Advisor | /smart-commit, /branch-cleanup |
| developer-toolkit | Essential development tools for code review, debugging, and refactoring | Code Reviewer, Debugger, Refactoring Architect | Code Review Practices | /quick-review, /debug-trace |
| security-toolkit | Comprehensive security toolkit for vulnerability scanning, secrets detection, and dependency auditing | Security Architect | Vulnerability Scanner, Secrets Detector, Dependency Audit | /security-scan |
| documentation-toolkit | Comprehensive documentation toolkit for README generation, API documentation, and changelog management | Docs Architect | README Generator, API Docs Generator, Changelog Manager | /generate-docs |
| code-quality-toolkit | Comprehensive code quality toolkit for detecting code smells, analyzing complexity, and planning refactoring | Quality Architect | Code Smell Detector, Complexity Analyzer, Refactoring Advisor | /quality-check |
| Plugin | Description | Agents | Skills | Commands |
|---|---|---|---|---|
| docker-toolkit | Docker containerization toolkit for image optimization and security scanning | Container Architect | Dockerfile Optimizer, Image Security Scanner | /optimize-image |
| python-toolkit | Python development toolkit with UV for fast package management and project structure | Python Expert | Python Packaging, Dependency Manager | /setup-project |
| react-toolkit | React development toolkit for component architecture and state management | React Architect | Component Patterns, State Management Advisor | /create-component |
| api-toolkit | API design toolkit for REST with endpoint design and OpenAPI documentation | API Architect | REST API Designer, API Documentation Generator | /generate-openapi |
| database-toolkit | Database design toolkit for schema modeling and query optimization | Database Architect | Schema Designer, Query Optimizer | /analyze-query |
| aws-toolkit | AWS cloud toolkit for service selection, cost optimization, and security | AWS Architect | AWS Cost Optimizer, Security Group Analyzer | /estimate-cost |
| terraform-toolkit | Complete Terraform IaC toolkit with module scaffolding and state management | Terraform Expert | Module Scaffolder, State Manager, Cost Estimator, Documentation Generator, Dependency Analyzer, Upgrade Assistant | - |
| kubernetes-toolkit | Kubernetes container orchestration toolkit for production-ready manifests | K8s Expert | Kubernetes Best Practices | - |
| nginx-toolkit | Nginx web server toolkit for load balancing, SSL/TLS configuration, and performance optimization | Config Architect | Load Balancer, SSL Helper | /optimize-nginx |
| cloudformation-toolkit | CloudFormation infrastructure as code toolkit for stack design and template validation | CFN Architect | Stack Designer, Template Validator | /validate-template |
| ansible-toolkit | Ansible automation toolkit for playbook design, role creation, and inventory management | Playbook Architect | Role Builder, Inventory Manager | /lint-playbook |
| azure-toolkit | Azure cloud toolkit for cost optimization, ARM templates, and infrastructure design | Azure Architect | Azure Cost Optimizer, ARM Template Helper | /estimate-azure-cost |
| gcp-toolkit | Google Cloud Platform toolkit for cost optimization, Cloud Build configuration, and infrastructure design | GCP Architect | GCP Cost Optimizer, Cloud Build Helper | /estimate-gcp-cost |
| dbt-toolkit | dbt data transformation toolkit for model building, testing, and documentation | DBT Architect | Model Builder, Test Generator | /validate-dbt |
| github-actions-toolkit | GitHub Actions CI/CD toolkit for workflow design, custom action creation, and matrix optimization | Workflow Architect | Action Builder, Matrix Optimizer | /optimize-workflow |
| react-native-toolkit | React Native mobile development toolkit for native module integration and performance optimization | RN Architect | Native Module Helper, Performance Optimizer | /create-native-module |
| nextjs-toolkit | Next.js framework toolkit for rendering strategies (SSR/SSG/ISR) and App Router patterns | Next.js Expert | SSR/SSG Advisor, App Router Helper | /create-page |
| go-toolkit | Go programming language toolkit for concurrency patterns, error handling, and idiomatic Go practices | Go Expert | Goroutine Patterns, Error Handling | /check-go |
| Plugin | Description | Agents | Skills | Commands |
|---|---|---|---|---|
| frontend-developer | Comprehensive frontend toolkit combining React, accessibility, performance, and responsive design | Frontend Lead | Accessibility Checker, Performance Optimizer, Responsive Design Advisor | /audit-accessibility |
| backend-developer | Comprehensive backend toolkit combining API design, database, security, and scalability | Backend Lead | API Security Checker, Database Migration Helper | /create-migration |
| devops-engineer | Comprehensive DevOps toolkit combining CI/CD pipelines, infrastructure automation, and monitoring | DevOps Lead | CI/CD Optimizer, Infrastructure Monitor | /check-pipeline |
| data-engineer | Comprehensive data engineering toolkit combining ETL pipelines, data quality, and data architecture | Data Architect | ETL Designer, Data Quality Checker | /validate-pipeline |
Agents make high-level architectural decisions and choose the right approach. They focus on the "what" and "why".
Example: The Code Reviewer agent decides review scope, prioritizes issues, and determines severity levels.
Skills handle specific operations with focused, step-by-step instructions. They focus on the "how".
Example: The Code Review Practices skill provides detailed guidance on conducting effective reviews.
Commands are user-invoked shortcuts for frequently-used prompts.
Example: /quick-review triggers a fast code review of specific files.
Agents automatically delegate to skills when needed. You get strategic thinking at the top level and tactical execution when required - all without manual intervention.
For automatic plugin installation across your team, add to your project's .claude/settings.json:
{
"extraKnownMarketplaces": {
"fastagent-marketplace": {
"source": {
"source": "github",
"repo": "armanzeroeight/agents-skills"
}
}
},
"enabledPlugins": [
"developer-toolkit@fastagent-marketplace",
"terraform-toolkit@fastagent-marketplace"
]
}When team members trust the repository, plugins install automatically.
Agents activate automatically based on context:
# Code review
> Review my recent changes
# Debugging
> Help me debug this error
# Infrastructure design
> Design a Terraform module for our VPCOr invoke explicitly:
> Use the code-reviewer agent to check my PR
> Ask the iac-architect agent about module designSkills are invoked automatically by agents or Claude when relevant:
# Triggers terraform-module-scaffolder skill
> Create a new Terraform module for S3 buckets
# Triggers code-review-practices skill
> What should I look for in this code review?Commands are user-invoked with slash syntax:
# Quick code review
/quick-review
# Debug trace
/debug-trace/plugin# Disable without uninstalling
/plugin disable <plugin-name>@fastagent-marketplace
# Re-enable
/plugin enable <plugin-name>@fastagent-marketplace/plugin uninstall <plugin-name>@fastagent-marketplace/plugin marketplace update fastagent-marketplaceOne of FastAgent's core principles is token efficiency. Here's how we help you save:
Instead of one massive plugin with everything, choose specific plugins:
# β Don't load everything
# One giant plugin with all technologies
# β
Load only what you use
/plugin install terraform-toolkit@fastagent-marketplace # Only if you use Terraform
/plugin install kubernetes-toolkit@fastagent-marketplace # Only if you use K8sSkills use progressive disclosure - detailed documentation is loaded only when needed:
skill/
βββ SKILL.md # Loaded always (concise)
βββ references/ # Loaded only when needed (detailed)
Each skill does one thing well, keeping context lean:
- β
terraform-state-manager- State operations only - β
terraform-cost-estimator- Cost analysis only - β
terraform-everything- Too broad, wastes tokens
We welcome contributions! Here's how:
- Fork the repository
- Create plugin directory:
plugins/your-plugin-name/ - Add agents, skills, and commands
- Update
.claude-plugin/marketplace.json - Submit a pull request
plugins/your-plugin/
βββ agents/
β βββ your-agent.md
βββ skills/
β βββ your-skill/
β βββ SKILL.md
βββ commands/
βββ your-command.md
- Agents: Focus on strategy and high-level decisions
- Skills: Provide tactical, step-by-step guidance
- Commands: Keep simple and focused on one action
- Documentation: Clear descriptions with trigger words
- Testing: Test with Claude Code before submitting
See Plugin Design Manifesto for detailed guidelines.
- Plugin Design Manifesto - Design principles and best practices
- Claude Plugins - Plugin system overview
- Claude Marketplaces - Marketplace management
- Claude Commands - Slash commands guide
- Claude Skills - Agent Skills documentation
- Claude Subagents - Subagents guide
- Issues: GitHub Issues
- Discussions: GitHub Discussions
- Website: gofastagent.com
MIT License - see LICENSE for details.
Made with π by the community, for the community.
